International Journal of Statistics and Applications

p-ISSN: 2168-5193    e-ISSN: 2168-5215

2018;  8(4): 203-210

doi:10.5923/j.statistics.20180804.05

 

Impact of Contraceptive Usage on Ideal Number of Children

Shilpi Rani Saha

Department of Statistics, Jahangirnagar University, Savar, Dhaka, Bangladesh

Correspondence to: Shilpi Rani Saha, Department of Statistics, Jahangirnagar University, Savar, Dhaka, Bangladesh.

Email:

Copyright © 2018 The Author(s). Published by Scientific & Academic Publishing.

This work is licensed under the Creative Commons Attribution International License (CC BY).
http://creativecommons.org/licenses/by/4.0/

Abstract

The present study investigates the relationship between ideal family size and contraceptive uses. The correlation of ideal family size are also studied and discussed. The data for the study was obtained from 1999-2000 BDHS which is a nationally representative sample. In the 1960s, advocates of population control often claimed pervasive demand for family planning services in the developing countries. Bangladeshi couples have accepted the small family norm. About 60 percent of women prefer a two-child family, and another more than 20 percent consider a three child family as ideal. Overall, the mean ideal family size among married women is 2.5 children and has not changed since 1993-1994 (1999-2000 BDHS). The logistic regression analysis was carried to identify the important variables which are influencing contraception. This analysis suggest that level of education, number of living children, place of residence, religion, electricity in locality of the respondent are important factors that influence contraceptive use of women. The analysis was designed to assess whether ideal number of children has any influence on current use of contraception. Although there exists a relationship between current use of contraception and desired family size, but its demographic impact will be small in view of the fact that most women had larger number of living children than the mean ideal children.

Keywords: Ideal Family, Contraceptive Use, Family planning, Logistic Regression, Demographic Analysis

Cite this paper: Shilpi Rani Saha, Impact of Contraceptive Usage on Ideal Number of Children, International Journal of Statistics and Applications, Vol. 8 No. 4, 2018, pp. 203-210. doi: 10.5923/j.statistics.20180804.05.

1. Introduction

Bangladesh is a country of about 157.8 million people (July 2017 estimated) [1], squeezed in an area of 147,570 sq. km. The age structure of the population is not favorable for the economic developments since about 24.76 percent of total population still remains below the age level of 15. This increases the dependency ratio and creates pressure on maternal child health (MCH), nutrition services, pre-school care and different levels of educational facilities etc. Population growth leads to an increase in the female population of reproductive ages [2, 3].
The population growth rate has declined from 2.44% per annum during 1961-81, 2.15% in 1991, 1.48% in 2001, 1.07% in (2017 est.), if held constant, will generate a doubling of population by 2050(According to BBS) [4, 5].
Some remarkable results were achieved in Bangladesh during last few decades may be termed as near miracle, as decline in the rates of mortality and fertility registered within a low socio-economic and low literacy scenario, challenge traditional theories of demographic transition [6]. Noticeable results reducing the total fertility rate from 6.4 to 3.3 in the nineties (UNFPA 2002) and 2.17 (2017 estimated) [7].
Low female age at marriage relatively high rates of neo natal, infant, child and maternal mortality, and increasing size of the elderly population. Population momentum, unwanted fertility and desire family size continue still at a higher level [8, 9]. Meanwhile the momentum achieved in the decline of TFR (2.10) has been leveled. Bangladesh achieved a very high level of CPR (62.4%) in 2014 according to World Bank [11-13].
According to BDHS 1999-2000 about 60% of women prefer a two child family and another more than 20% consider a three child family ideal. Overall the mean ideal family size among married women is 2.5 children and has not changed since 1993-1994 [10, 14, 15].
Fertility decline achieved by mid-nineties is remarkable but since the TFR has leveled [16]. The main mechanism of fertility decline is increased use of contraceptives achieved particularly through female method of contraception (oral contraceptive) DHS survey demonstrated emergence of fertility differentials [7, 17, 18]. Contraceptive was considered as one of the control concepts for reducing maternal and child mortality [19].
Contraceptive prevalence has steadily grown in Bangladesh since 1975. In 1975, only 8 percent of currently married women reported using a family planning method, compared with 62 percent in 2014, a sevenfold increase in the contraceptive prevalence rate (CPR) has experienced in Bangladesh in less than 40 years for any method [20]. A recent study found contraceptive prevalence rate is 81.27 percent [21].
Ideal family size indicates some kinds of preference of the women where they want to stop their fertility [6, 17]. More of the ideal family size perhaps reveals the feature that they were not used to use contraceptive to limit their children upto the limit of ideal number of children. Some factors may be associated with this problem [9]. The data indicates that only about 32 percent (31.9 percent) of the total women already exceed their ideal family size. About 68 percent of the women do not reach or exceed their family size. Relatively they are young and duration of marriage are shorter than the others. Among them 60.5 percent do not use contraceptive. In reality those women do not exceed or reach the ideal family size are likely to use contraceptive than those who exceeded their ideal family size.
As per literature survey it was found that more studies were required for effective action in this area. (Haq et al. [22] reported that, contraceptive use decreases with age, indicating older women were more reluctant to use contraceptive than their younger counterparts. To this view point more studies were done. In this work, author collected secondary data from BBS Dhaka and analyzed in different statistical process to explain overall findings. In this paper an attempt has been made to examine the contraceptive use status of the Bangladeshi married women who do not reach or exceed their ideal family size until yet.

2. Materials and Methods

2.1. Data Sources

The data from the Bangladesh Demographic and Health Survey (BDHS) 1999-2000 are used in this study. The detailed methodology and survey design were explained in the report [15].
A total of 10,268 households were selected for the sample, of which 9,854 were successfully interviewed. The shortfall is primarily due to dwelling that were vacant or in which the inhabitants had left for an extended period at the time they were visited by the interviewing teams of the 9,922 households occupied. 99 percent were successfully interviewed in these households, 10,885 women were identified as eligible for the individual interview (i.e., ever-married and age 10-49) and interviews were completed for 10,544 or 97 percent of them.
The BDHS data entry and editing programs were written in ISSA (Integrated System for Survey and Analysis).Data processing commenced in mid-December 1999 and was completed by end of April 2000 [15].

2.2. Statistical Methods

Frequency distribution was used to describe to show current use of contraceptive methods between the two groups of women of aged less 25 who did not exceed and exceeded ideal family size. Which factors are associated with contraceptive use of married women are also examined using statistical methods. The logistic regression is also applied in this study to examine the differentials of contraceptive use among the categories of each independent variable. Only the odds ratios and corresponding 95 percent confidence Intervals (CI) are presented with explanations. The Statistical Package for Social Science (SPSS) software was used for data analysis.

2.3. Logistic Regression Model

Examination of each independent variable individually, as it is done bi-variate analysis, can only provide preliminary idea about the importance of each variable is itself. To examine the relative importance of all variables simultaneously, multivariate techniques are needed. Multiple regression analysis is a related technique that quickly comes to mind. However this technique becomes difficult when dependent variable can have only two values, an event occurring or not occurring. When dependent can have only two values the assumptions necessary for hypothesis testing in regression analysis are necessarily violated. For example it is not reasonable that the predictor values cannot be interpreted as probabilities. They are not constrained to fall in the interval between 0 and 1.
Regression methods have become an integral component of any data analysis concerned with describing the relation between a response variable and one or more explanatory variables. It is often the case that the outcome variable is discrete, taking one or two or more possible values. Over the last decade the logistic regression model has become, in many fields, the standard method of analysis in this situation. What distinguishes a logistic regression from the linear regression model is that the outcome variable in logistic regression analysis is binary or dichotomous.
The difference between logistic and linear regression is reflected both in the choice of a parameter model an in the assumptions. Once this difference is account for, the methods employ in an analysis using the logistic regression follow the same general principle used in linear regression.
Logistic regression model does not require any distribution assumptions concerning explanatory variables (Cox, 1970). The logistic regression model can be used not only to identify risk factors but also predict the probability of success. The general logistic model expresses a qualitative dependent variable as a function of several independent variables, both qualitative and quantitative (Fox, 1984). It is important to understand that the goal of logistic regression is the same as that of any model building technique used in statistics to find the best fitting and most parsimonious, yet reasonable model to describe the relationship between and outcome (dependent and response variable) and a set of independent (predictor or explanatory) variables. The independent variables are often called covariates.
Suppose there are n individuals, some of them are called “success” and other are called “failure”. Let Yi denotes the dependent variable for ith observation and Yi=1 if the ith individuals is success and Yi=0 if the ith individuals is a failure. Consider the collection of P independent variables which will be denoted by the vector X’ = (x1,x2,………….,xp) and the vector of the coefficient of X is β’= (β1, β2,β3,……………,βp) these variables are either qualitative, such as religion, residence or quantitative, such as number of living children. For the moment let us assume that each of these variables at least interval scaled. Let the conditional probability that the outcome present is denoted by Then the logit of the multiple logistic regression model is given by the equation.
(1)
In which case
(2)
And
(3)
Equation (2) and (3) look complicated, however the logarithm of the ratio π(Y) and 1-π(Y)
Which of π(Y), turns out to be a simple linear function of X. That is,
(4)
Which express the log odds of the occurrence on an event as a linear function of independent variables. The logit is thus the logarithm of odds of success, that is, the logarithm of the ratio of the probability of success to the probability of failure. It is also called the logit transformation of π(Y) and equation (4) is a linear logistic model. It has several special properties: π(Y) is bounded only between 0 and 1. If π(Y)<0.5, logit π(Y) is negative: while π(Y)>0.5, logit π(Y) is positive.
In logistic regression parameter of the models are estimated by maximum likelihood method. That is the coefficient that makes observed result most “likely” are selected. The contribution of independent variable in logistic regression dependent on the other individual variables and the interpretation is difficult when are highly correlated. In logistic regression, just as linear regression, the codes for the independent variable must be meaningful. The values for the independent variables must be decoded by the creating a new set of variables that corresponds, in some way, the original categories.

3. Results

3.1. Current Method

Works on ideal family size preference essentially include the topics on perception, attitude and practices of contraceptive methods. Knowledge of family planning method is high among Bangladeshi couple [23]. Virtually all respondent of age group less than 25 know at least one modern method of family planning, 60.5 percent women of aged less 25 who are not using any contraceptive method did not exceed ideal family size. And 41 percent women using contraceptive method and exceeded ideal family size. 65.6 percent women of aged greater than 25 who exceeded ideal family size using contraceptive method. This confirms the finding documented in the earlier BDHS survey that younger women have begun to appreciate the advantages of deliberately controlling childbirth early in marriage. Since 1993-1994, over a period of six years, contraceptive use has increased among women 15-19 by nearly 50 percent, from 25 to 38 percent in the 1999-2000 BDHS survey.
Table 3.1.a. Status of current contraceptive use of the women aged below 25
     
Table 3.1b. Status of current contraceptive use of the women aged above 25
     

3.2. Desire for Future Children

Bangladesh has the highest preference for sons over daughters. Sons are generally preferred over daughters owing to a complex interplay of economic and socio-cultural factors. Sons contribute more than daughter to family income, provide adequate support in old age to their parents, impose less of a financial burden and carry forward the family name (Nag 1991: Ali, 1989) [10, 24].
The desire for additional children declined noticeably in over the past decade. Table 3.2a shows that the desire for future children of the women aged less than 25 who did not exceed and exceeded ideal family size and whose ideal family size equal desire family size. 89.5 percent women who did not exceed ideal family size want more children. 83.7 percent women who exceeded ideal family size want no more children. The women who did not exceed ideal family size and wants more children their mean number of ideal children is 2.4. It is greater than those women who exceeded ideal family size. Among younger (<25) women wants no more children (7.6 percent) is small who did not exceed ideal family and large (83.7 percent) who exceeded ideal family size.
Table 3.2.a. Desire for future children of the women aged below 25
     
Table 3.2.b. Desire for future children of the women aged above 25
     
Among the women aged greater than 25 a large number of women (95.6 percent who exceeded ideal family size wants no more children their mean number of ideal children is 2.2. Mean number of ideal children is highest 3.5 who did not exceed ideal family size and who are undecided to use contraception.
It is seen that there is strong preference for sons prevail in Bangladesh. In addition, among couples with more than one child, the proportion wanting no more children was higher among those with at least one son, compared those having daughters only.

3.3. Current Contraceptive Method Use

In the contraceptive prevalence survey current use is defined as those “now using” a family planning method. That is, any respondent reporting “using a family planning method” in reply to the following question is counted as a current user. “Are you now using any family planning method? and (if yes),”which method are you using” [25, 26].
The result show significant increase in current use of contraception methods between the two groups of women aged less than 25 who did not exceed and exceeded ideal family size. Using modern method increased (51.2 percent) of those women who exceeded ideal family size and age is greater than 25.
Table 3.3.a. Method of current contraceptive using of the women aged above 25
     
Table 3.3.b. Method of current contraceptive using of the women aged below 25
     

3.4. Ideal Children

The attempt to measure ideal family size is somewhat hazardous and includes concepts of “wantedness” of most recently born child; of whether or not the couple wants an additional child either sooner, later, or never, and ideal family size at the time of marriage-theoretically not considering how many children the couple already has. (1999-2000 BDHS) [15].
The women of aged less than 25 who did not exceed ideal family size in fact most women (45.1 percent) show a strong desire to have at least one boy and at least one girls as ideal and their mean number of ideal children 2.0 and it increased (78 percent) who exceeded ideal family size.
Table 3.4 b present the women who did not exceed ideal family size (23.7 percent) required 1 boy and 1 girl as ideal and it is increases (57.6 percent) women who exceeded ideal family size.2 boys and 1 girls is ideal among women who exceeded or did not exceed ideal family size and mean number of ideal children in both the group. It is also significant at p<0.00 percent level.
Table 3.4.a. Number of ideal children of the women aged below 25
     
Table 3.4.b. Number of ideal children of the women aged above 25
     

3.5. Factors Influencing Desire for Ideal Children

In order to identify the factor that is there ideal family size were investigate by applying logistic regression model.
There is a strong argument that socio-economic and demographic differentials like education, religion, access to information, residence etc. influence contraceptive practices of women. To justify this argument from survey evidence we like to apply a powerful statistical technique like logistic analysis. Logistic analysis is nothing but a regression technique dealing with criterion variables. As we are doing the study to see actual performances of contraceptive practices by women in Bangladesh, we put emphasis to find out which factors are significantly influencing contraceptive practice of women with the help of logistic regression technique. Along with this we will explain the model of ideal family size have any impact on current contraceptive use.
Before going into the explaining of the models we prefer to get a brief idea about the methodology of logistic regression.

3.6. Findings and Interpretation of the Result

Application and interpretation of logistic regression technique is different from usual regression technique. Fitting logistic regression requires categorization of variables. Our dependent variable is the current use of contraception. It is a dichotomous variable. Those who are currently using a method considered 1 otherwise 0. The independent variables included in the model are education, religion, place of residence, access to information, number of living children etc.
The variables are defined below
The regression equation is
Y = .6906 -.0797X1 + .3167x2 + .3856x3 + .4206x4 + .5735x5 - .0801x6 + .2148x7 - .3806x8 + 1.0544x9
From table 3.5, we observed that literacy, access to information through newspaper, radio, TV, Religion, number of living children are found strongly significant in explaining factors influencing desire for ideal children. The variables like to type of place of residence, highest education level, possess electricity seems to be insignificant.
Table 3.5. Logistic regression models to explain the factors influencing desire for ideal children of Bangladeshi women
     
From the above model we observe that among all of the explanatory variables number of living children is the most powerful determinant in explaining the factors influencing desire for ideal children. The standardized regression coefficient for the variable, respondent’s number of living children is 1.0544, which is positively related with the dependent variable current use status of the respondent. The result indicates that if number of living children increases, the contraceptive uses will also increases. The next logistic coefficient is .3856 for the literacy of the respondents. Because educated women are exposed to family planning and this might have contributed to the current using contraception. The next important determinant is religion of the respondents and the regression coefficient is -0.3806 which indicates that Muslim women are found to be significantly lower use of contraceptive than non-Muslim. Regression coefficient is -0.0801 of possess electricity which indicate that contraceptive use is higher where electricity is available and another regression coefficient of type of place of residence is -.0797 indicate that urban women are more exposure to use in contraception than rural women.
Access to information is also important variable that rises women’s awareness and help to make decision regarding contraceptive use thereby accelerate the pace of contraceptive uses. Newspaper, radio and TV are powerful media that play key role in increasing the participation in contraceptive practices. Watches TV every week give the most significant result for the current use of contraception of those women who exceeded ideal family size.
Table 3.5 Also shows the odds ratios, which are easier to interpret. For example, the odds ratio of 1.0544 is 2.8702 for the number of living children of the respondents, means that contraceptive use is three times higher for those women who have three or more children than those whose are childless, when the other independent variables are held constant.

4. Discussion

The study covered two areas of ideal family size preference of women and the impact of ideal number of children of current contraceptive use. One is about the women who did not exceed or exceeded ideal family size and the other is about the contraception practice of those women. And some selected background of those who exceeded or did not exceed: mean number of living children, education, access to mass media, electricity, urban-Rural etc. Factor influencing ideal family size also investigated.
For this we have applied a statistical tool like logistic regression to identify socio-economic and demographic differentials of contraceptive practice and ideal family size preference.
The percentage distribution of the ideal children of the respondents show that among younger (<25) women 1 boy and 1 girl as ideal from the maximum women who did not exceed (45.1 percent) or exceeded (78 percent) ideal family size and the figure is (23.7 percent) and (57.6 percent) whose aged greater than 25.
Our final and much emphasized issue was to identify the factors that have influence on contraceptive practice of women. We used univariate analysis to assess the association between the variables. The x2 values provided the strata of relationship. These significance variables were considered in the logistic regression analysis to determine the statistically significance variables. The logistic regression analysis shows that education is highly significant to explain contraceptive use of women. The positive sign of the regression coefficient of this variable indicate that this variable is positively influencing the use of contraceptives among the Bangladeshi women, This means that with the increase of the education level the use of contraceptive increases. Similar interpretation can be inferred in case of residence to explain contraceptive use of women. Access to information through newspaper, radio, TV etc. is an important determinant to grow awareness among the users about contraception. But we didn’t found any significance of those media to explain contraceptive use of women aged less than 25 except in the model to explain contraceptive use of women aged greater than 25. In this model, we observed that TV is found significant to influence decision regarding contraceptive use. So, it is expecting that media should play more active and lively role to make awareness about contraception and thereby to increase participation rate. The availability of electricity in the locality is another variable, which affects contraceptive use. This study indicates that the use of contraceptive is higher who did not exceed or exceeded ideal family size where electricity is available. Religion is another factor, which is found negatively significant in case of women aged greater than 25 who exceeded ideal family size. The number of living children is negatively significant of the women who did not exceed ideal family size. This means that with increased number of living children and decreases the use contraceptive methods.

5. Conclusions

The major conclusion of this study is that although the desired family size has declined during the past decades or so, it still remains above three children. This is partly because the current stated mean ideal number of children (2.5 children) is still high. Unless high mean ideal family size preference is reduced, the prospect for fertility decline and hence to achieve replacement fertility will be a great concern to the policy makers. The ideal family size demonstrates that women do not need to convince of the value of smaller families. Thus, priority must be given to providing quality family planning services that enable women to adopt and continue the contraceptive method that is right for them.
Since mean ideal children is still high and contraceptive use is also high among couples who exceeded, the demographic impact is expected to be low. On the other hand women aged 25 years and above their contraceptive use are higher than their counterparts. Logistic regression analysis also supported that contraceptive is an important determinant of ideal number of children. The study does not answer to the question that the relationship between contraceptive and ideal desired family size although positive will not affect demographic objectives.

References

[1]  Central Intelligence Agency World FactBook (January 20, 2018).
[2]  Caldwell J.C. Number 3, September 1999, The delayed Western fertility decline in English-speaking countries. Population and Development Review, Volume 25.
[3]  Population and Development: post-ICPD, June, 1999, Achievements and challenges in Bangladesh, Ministry of Health and Family welfare.
[4]  Population Census 2001, August 2001, Preliminary Report, Bangladesh Bureau of Statistics, Statistics Division, Ministry of planning.
[5]  Mid-Term Review (MTR), November 2000, Annual Program Review (APR), Bangladesh Health and population Sector Program (HPSP) 1998-2003, Mission Term Report (15 October – 3 November 2000).
[6]  Study on Urban Adolescents: July 2002, Knowledge, Attitude and practice related to reproductive and sexual Health in Bangladesh, Save the children, UK.
[7]  Report on the international conference on population and development, September 1994, United Nation – New York, 1995, Cario, 5-13.
[8]  Haider S.J., 2nd July 2001, Assesment of Bangladesh Health System Performance as a Measre to alleviate Poverty and Improve Quality of Life, Partnership Agreement on Poverty Reduction Between ADB and GOB, Research Evaluation Associates for Development(READ).
[9]  Bangladesh: Progress in Poverty Reduction, The independent, Tuesday, 1 October 2002, Background Paper, Bangladesh Development Forum, Paris, March 13-185, 2002.
[10]  Health Features in Bangladesh: Some Key Issues and Options.
[11]  Celand. J, Phillips J.F, Amins. S, Kamal. J.M, 1994, The determinants of reproductive change in Bangladesh, Success in A Challenging Environment. The world bank Regional and sect oral studies.
[12]  Bangladesh 2020, The university Press Limited, 1998, A Long-run Perspective Study, Bangladesh Development Series, The world Bank and Bangladesh Center for Advanced Studies.
[13]  The World Bank, Washington D.C., 1994. Population and Development, development in Practice, Implications for the World Bank.
[14]  Haider S. J., December 1997, Behavior Change Communication Strategies and Action Plan for HAPP-5: 1998-2003, Ministry of Health and Family Welfare And CIDA Monitoring and Technical Advisory Unit.
[15]  Bangladesh Demographic and Health Survey 1999-2000, May 2001, NIPORT, MItra and Associates, ORC Macro.
[16]  Kabir M. & Chowdhury Md. A.A.M., Plateauing Fertility in Bangladesh: What are its Correlates and Investigation of Proximate Determinants?, First Draft.
[17]  Brkat. A. Howlader. S.J. MOHFW, 1997, Population and development issues in Bangladesh National Plan of Action Based on ICPD’94 Recommendation.
[18]  Health and population Sector Program 1998-2003, April 1998, HPSP< Program Implementation Plan, Part – 1, Ministry of Health and Family Welfare, Government of the people’s Republic of Bangladesh.
[19]  Akkul Chandra Das (2015), Contraceptive using Trends in Bangladesh. Int. J. Bus. Sci. Res. 3(3): 184-188 Retrieve from http:// .
[20]  Fauzia Akhter Huda, Yolande Robertson, Sabiha Chowdhuri, Bidhan Krishna Sarkar, Laura Reichenbach and Ratana Somrongthong, Reproductive Health 2017, “Contraceptive Practices among married women of reproductive age in Bangladesh: a review of the evidence”. .
[21]  Al kibria, Gulan Muhammad: Hossen, Shakir, Barsha, Rifatharaalam, Sharmeen, Atia: Paul, Sujit Kumar: Uddin, S.M. Iftekhar (2016). “Factorsaffectings contraceptive use among married womrn of reproductive age in Bangladesh. Journal of Molecular studies and Medicine Research. 2(1): 70-79. Dol:10. 18801/jmsmsr 020116.09.
[22]  Iqramul Haq, Saifullah Sakib and Ashis Talukder. Dec.1, 2017, Sociodemographic Factors on Contraceptive use among Ever-Married Women of Reproductive Age: Evidence from three Demographic and Health Survey in Bangladesh.
[23]  Ahmed, Kapil, Rahman, Mizanur Giinneken Van Jeroen. 1998. “Induced abortion in Matlab, Bangladesh: trends and determinants” .International Family Planning Perspective, 24(3): 128-132.
[24]  Das, N. 1987. Sex Preference and Fertility Behavior. A Studt of Recent Indian Data. Demography 24(4): 517-530.
[25]  UNO. 2001. Levels and Trends of Contraceptive Use as Assessed in 1998. Department of Economic and Social Affairs. Population Division. New York.
[26]  Haider S.J., Streatifield K. Ph.D., Karim M. A. 1995, comprehensive guidebook to Bangladesh family planning MCH program, Research evaluation for development (READ), The Population Council and Bangladesh Ministry of Health and Family Welfare.